To overcome the problem that soft sensor models cannot be updated with the process changes, a soft sensor modeling algorithm based on hybrid fuzzy c-means （FCM） algorithm and incremental support vector machines （I...To overcome the problem that soft sensor models cannot be updated with the process changes, a soft sensor modeling algorithm based on hybrid fuzzy c-means （FCM） algorithm and incremental support vector machines （ISVM） is proposed. This hybrid algorithm FCMISVM includes three parts： samples clustering based on FCM algorithm, learning algorithm based on ISVM, and heuristic sample displacement method. In the training process, the training samples are first clustered by the FCM algorithm, and then by training each clustering with the SVM algorithm, a sub-model is built to each clustering. In the predicting process, when an incremental sample that represents new operation information is introduced in the model, the fuzzy membership function of the sample to each clustering is first computed by the FCM algorithm. Then, a corresponding SVM sub-model of the clustering with the largest fuzzy membership function is used to predict and perform incremental learning so the model can be updated on-line. An old sample chosen by heuristic sample displacement method is then discarded from the sub-model to control the size of the working set. The proposed method is applied to predict the p-xylene （PX） purity in the adsorption separation process. Simulation results indicate that the proposed method actually increases the model＇s adaptive abilities to various operation conditions and improves its generalization capability.展开更多
In order to directly separate trivalent minor actinides （MA： Am, Cm） from fission products （FP） containing rare earths （RE） in high level radioactive liquid waste （HLLW）, the authors have challenged to develo...In order to directly separate trivalent minor actinides （MA： Am, Cm） from fission products （FP） containing rare earths （RE） in high level radioactive liquid waste （HLLW）, the authors have challenged to develop a simplified MA separation process by extraction chromatography using a single column. Attention has been paid to a new type of nitrogen-donor ligands, R-BTP （2,6-bis（5,6-dialkyl-1,2,4-triazin-3-yl） pyridine, R： alkyl group） as an extractant because it shows high extraction selectivity for Am（Ⅲ） over RE（Ⅲ）. It is known that the R-BTP ligands show different properties such as adsorbability and stability by hav- ing different alkyl groups. Therefore, some novel adsorbents were prepared by impregnating different types of R-BTP ligands （isohexyl-, isoheptyl- and cyheptyl-BTP） and a similar ligand to the R-BTP, ATP （2,6-bis（l-aryl-lH-tetrazol-5-yl）pyridines）, into the porous silica/polymer support （SiOrP particles）. This work deals with comparison in adsorption and desorption prop- erties of Am and some FP in HNO3 solution onto such R-BTP type adsorbents, as well as chemical and radiolytic stability of the adsorbents. Then the possibility of a single-column separation of MA from main FP was pursued by evaluating the results of column experiments using the most promising adsorbent （isohexyl-BTP/SiO2-P） under temperature control. In addition, elu- tion behaviors of U and Pd were also estimated.展开更多
基金Supported by the National Natural Science Foundation of China （60421002） and priority supported financially by ＂the New Century 151 Talent Project＂ of Zhejiang Province.
文摘To overcome the problem that soft sensor models cannot be updated with the process changes, a soft sensor modeling algorithm based on hybrid fuzzy c-means （FCM） algorithm and incremental support vector machines （ISVM） is proposed. This hybrid algorithm FCMISVM includes three parts： samples clustering based on FCM algorithm, learning algorithm based on ISVM, and heuristic sample displacement method. In the training process, the training samples are first clustered by the FCM algorithm, and then by training each clustering with the SVM algorithm, a sub-model is built to each clustering. In the predicting process, when an incremental sample that represents new operation information is introduced in the model, the fuzzy membership function of the sample to each clustering is first computed by the FCM algorithm. Then, a corresponding SVM sub-model of the clustering with the largest fuzzy membership function is used to predict and perform incremental learning so the model can be updated on-line. An old sample chosen by heuristic sample displacement method is then discarded from the sub-model to control the size of the working set. The proposed method is applied to predict the p-xylene （PX） purity in the adsorption separation process. Simulation results indicate that the proposed method actually increases the model＇s adaptive abilities to various operation conditions and improves its generalization capability.
文摘In order to directly separate trivalent minor actinides （MA： Am, Cm） from fission products （FP） containing rare earths （RE） in high level radioactive liquid waste （HLLW）, the authors have challenged to develop a simplified MA separation process by extraction chromatography using a single column. Attention has been paid to a new type of nitrogen-donor ligands, R-BTP （2,6-bis（5,6-dialkyl-1,2,4-triazin-3-yl） pyridine, R： alkyl group） as an extractant because it shows high extraction selectivity for Am（Ⅲ） over RE（Ⅲ）. It is known that the R-BTP ligands show different properties such as adsorbability and stability by hav- ing different alkyl groups. Therefore, some novel adsorbents were prepared by impregnating different types of R-BTP ligands （isohexyl-, isoheptyl- and cyheptyl-BTP） and a similar ligand to the R-BTP, ATP （2,6-bis（l-aryl-lH-tetrazol-5-yl）pyridines）, into the porous silica/polymer support （SiOrP particles）. This work deals with comparison in adsorption and desorption prop- erties of Am and some FP in HNO3 solution onto such R-BTP type adsorbents, as well as chemical and radiolytic stability of the adsorbents. Then the possibility of a single-column separation of MA from main FP was pursued by evaluating the results of column experiments using the most promising adsorbent （isohexyl-BTP/SiO2-P） under temperature control. In addition, elu- tion behaviors of U and Pd were also estimated.
